**Salivary Levels of Titanium, Nickel, Vanadium, and Arsenic in Patients Treated with Dental Implants: A Case-Control Study**

**Piero Papi 1, Andrea Raco 1, Nicola Pranno 1,\*, Bianca Di Murro 1, Pier Carmine Passarelli 2, Antonio D'Addona 2, Giorgio Pompa 1,**† **and Maurizio Barbieri 3,**†


Received: 22 March 2020; Accepted: 23 April 2020; Published: 27 April 2020

**Abstract:** Background: Recent articles have hypothesized a possible correlation between dental implants dissolution products and peri-implantitis. The null hypothesis tested in this case-control study was that there would be no differences in salivary concentrations of titanium (Ti), vanadium (V), nickel (Ni) and arsenic (As) ions among patients with dental implants, healthy (Group A) or affected by peri-implantitis (Group B), compared to subjects without implants and/or metallic prosthetic restorations (Group C). Methods: Inductively coupled plasma mass spectrometry was used to analyze saliva samples. One-way repeated-measure analysis of variance (ANOVA) was used to identify statistically significant differences in the salivary level of Ti, V, Ni and As between the three groups. Results: A total of 100 patients were enrolled in the study (42 males and 58 females), distributed in three groups: 50 patients in Group C, 26 patients in Group B and 24 patients Group B. In our study, concentrations of metallic ions were higher in Group A and B, compared to the control group, with the exception of vanadium. However, there were no statistically significant differences (*p* > 0.05) for metallic ions concentrations between Group A and Group B. Conclusions: Based on our results, there are no differences in titanium or other metals concentrations in saliva of patients with healthy or diseased implants.

**Keywords:** dental implants; saliva; corrosion; titanium; metallic ions

#### **1. Introduction**

Dental implants are usually made of commercially pure titanium or titanium-based alloys (Ti-6Al-4V), these metals, as well as showing great long-term success and survival rates [1,2], are known to be bio compatibles and chemically inert in the oral cavity and consequently considered corrosion-resistant. Corrosion is a state of metal's deterioration caused by oxidation or chemical action, thwarted by a titanium dioxide layer (TiO2), spontaneously covering the implant surface in presence of oxygen and providing great resistance and stability to the implant [3,4]. Nevertheless, titanium alloys can be affected by this gradual degradation and the structural and mechanical integrity of the

implant may be jeopardized and at risk of failure. Corrosion leads to titanium dissolution, release and dispersion of metal ions and particles into soft and hard tissues either, which may be a possible factor in peri-implant inflammatory processes [5,6].

The factors that trigger corrosion are acidification of the oral environment or the release of lactic acid by bacteria, such as *Streptococcus mutans*, and also chemical stimuli by fluorides (with a concentration range greater than 100 ppm) normally present in home and professional oral care products [7–9].

Titanium and other metals particles released by dental implants are not totally bioinert: they induce the release of mediating inflammation cytokines, such as tumor necrosis factor-alpha (TNF-α), interleukin 1 beta (IL-1β) and the secretion of RANKL. These present an immunogenic potential which—by acting as a secondary inflammatory stimulus in peri-implantitis—amplifies bone resorption [10]

Under physiological conditions, osteoprotegerin (OPG) secreted by the fibroblasts of the periodontal ligament and osteoblasts, subtracts the receptor activator of nuclear factor kappa-B ligand (RANKL) from the link with its true receptor activator of nuclear factor kappa-B (RANK), creating a real competitive inhibition, preventing differentiation from the pre-osteoclast into osteoclasts and subsequent bone resorption [11,12]. In the presence of inflammation, an up-regulation of RANKL is associated with a down-regulation of OPG, which is produced in minimal quantities. The RANKL produced will, therefore, bind the RANK present on the osteoclastic precursors, with consequent massive bone resorption [13,14]

Furthermore, other kinds of corrosion can occur, such as mechanical corrosion due to functional stresses of implants or the cracking or weakening of its prosthetic components [7,8].

Once ultimately dispersed in the peri-implant environment, metal ions and particles are no longer biocompatible and inert, and are, therefore considered a potential incentive to the inflammatory process in peri-implantitis [5]. A study in mice immune cells has demonstrated that Titanium ions release pro-inflammatory cytokines involved in bone resorption, demonstrating that peri-implantitis may be directly related to corrosion and, therefore, with metal ions dispersion [15].

Peri-implantitis treatment and risk factors are still controversial [16,17], with incontrovertible evidence just for periodontitis, inadequate plaque control and lack of supportive periodontal therapy [18–20].

Furthermore, smoking, excess cement and other systemic conditions have all been described as potential risk indicators, however there is inconclusive evidence to draft clear conclusions [18,21–26].

Recent articles have hypothesized a possible correlation between titanium dissolution products and peri-implantitis [27–29] therefore, the aims of this case-control study were to evaluate salivary metallic ions dissolution in patients with dental implants, healthy or affected by peri-implantitis.

The null hypothesis tested was that there would be no differences in the salivary concentrations of titanium (Ti), vanadium (V), nickel (Ni) and arsenic (As) ions among patients with dental implants, healthy or affected by peri-implantitis, compared to subjects without implants and/or metallic prosthetic restorations.

#### **2. Material and Methods**

#### *2.1. Study Design*

To address the research purpose, the authors designed a case-control study, conducted at the Departments of Oral and Maxillo-Facial Sciences, at "Sapienza" University of Rome.

#### *2.2. Study Population*

From May 2018 to May 2019, all subjects who underwent previous implant surgery since January 2008 at the Oral Surgery Unit, Policlinico Umberto I, "Sapienza" University of Rome were identified and consecutively evaluated.

#### *J. Clin. Med.* **2020**, *9*, 1264

Each patient received detailed descriptions of the study protocol and all subjects signed an informed consent form to be included in the study population, according to the World Medical Declaration of Helsinki. The Institution Review Board of "Sapienza" approved the study (Ref. 4939/2018).

Eligible patients were divided into two groups: subjects with clinically healthy implants (Group A) and patients with peri-implantitis (Group B).

Patients in group A were enrolled in the study based on the following inclusion and exclusion criteria:


Patients in Group B had to meet the following inclusion and exclusion criteria:


Furthermore, a third group of patients (Group C) without dental implants, derived from a population of subjects attending the Oral Surgery Unit from May 2018 to May 2019 for wisdom tooth removal, was enrolled in accordance with the following inclusion and exclusion criteria:


Patients' data collected included sex and age.

For each implant, the following clinical measurements were recorded at six sites per implant by using a periodontal probe (PCP-Unc 15, Hu-Friedy®, Chicago, IL, USA):


#### *2.4. Radiographic Assessment*

Mesial and distal implant crestal bone levels were measured on standardized (Rinn Centratore XCP Evolution 2003, Dentsply, Rome, Italy) digital periapical X-rays for each implant using a calibrated software (SOPRO Imaging, Acteon Group, Norwich, UK). The implant length and width were used as references for calibration of measurements, with the bone level digitally evaluated by measuring the distance between the implant shoulder and the first visible bone contact on the implant.

Case definitions for epidemiological or disease surveillance studies of the 2017 World Workshop on the Classification of Periodontal and Peri-Implant Diseases and Conditions were adopted to establish the diagnosis of peri-implant diseases [30].

#### *2.5. Saliva Collection*

Five milliliters of saliva were collected from each patient using the unstimulated drainage method, or alternatively by leaving the saliva flow passively from the lower lip directly into designated sterile tubes.

Patients were advised not to take food or drink and to avoid oral hygiene procedures before saliva collection.

All samples were stored at a temperature of 1 ◦C, in order to block any biologic process still in place.

#### *2.6. ICP-MS Analysis*

Samples were analyzed using a Thermo Scientific XSERIES2 ICP-MS instrument (Thermo-Fisher Scientific, Waltham, MA, USA) at the Department of Earth Sciences at "Sapienza" University of Rome. Samples were processed as previously described [29]. The concentrations of Ti, V, Ni and As were quantified in μg/L.

#### *2.7. Statistical Analysis*

The required sample size was calculated using statistics software (GPower 3.1.9.2, Heinrich-Heine-Universität, Düsseldorf, Germany). A power analysis using the repeated measures ANOVA with three measurements, an alpha level of 0.05 and a medium effect size (f = 0.56) showed that 100 subjects would be adequate to obtain 95% power in detecting a statistical difference in the salivary level of Ti, assuming a loss to follow-up of 20% [31].

A database was created using Excel (Microsoft, Redmond, WA, USA). Descriptive statistics were calculated for each variable. The Shapiro–Wilk test was used to determine whether or not the data conformed to a normal distribution.

One-way repeated-measures ANOVA was used to identify statistically significant differences in the salivary level of Ti, V, Ni and As between three different groups: Control group, patients with healthy implant and patients with implant affected by peri-implantitis. Pairwise Tukey honestly significant difference (HSD) test for multiple comparisons.

The chi-squared test of homogeneity was used to evaluate the presence of differences in the proportion of males and females among the three groups.

Data were evaluated using standard statistical analysis software (version 20.0, Statistical Package for the Social Sciences, IBM Corporation, Armonk, NY, USA). A *p* value <0.05 was considered as statistically significant.

#### **3. Results**

A total of 100 patients were enrolled in the study (42 males and 58 females; age 49.02 ± 11.37 years). The patients were thus distributed in the three groups: 50 patients in Group C, 26 patients in Group B and 24 patients Group B. There were no significant differences in the proportions of males and females in the three groups (*p* = 0.620). Patients' demographics are reported in Table 1.



SD, standard deviation; PPD, probing pocket depth; MBL, marginal bone loss; NA, not applicable.

#### *3.1. Titanium*

The mean concentration of Ti was 136.65 ± 263.28 μg/L in the Group C, 489.60 ± 227.86 μg/L in the Group A and 492.83 ± 313.90 in the Group B. Statistical significant differences in mean concentration were found between the Group C and the Group A (*p* < 0.001) and between the Group C and the Group B (*p* < 0.001) (Table 2).

#### *3.2. Nickel*

The mean concentration of Ni was 4.77 ± 8.33 μg/L in the Group C, 24.99 ± 12.47 μg/L in the Group A and 23.50 ± 10.12 in the Group B. Statistical significant differences in mean concentration were found between the Group C and the Group A (*p* < 0.001) and between the Group C and Group B (*p* < 0.001) (Table 2).

#### *3.3. Vanadium*

The mean concentration of V was 1.30 ± 4.28 μg/L in the Group C, 2.27 ± 2.43 μg/L in the Group A and 2.01 ± 1.35 in the Group B. The group means of V concentration were not statistically significant different (*p* = 0.440) (Table 2).

#### *3.4. Arsenic*

The mean concentration of As was 0.01 ± 0.01 μg/L in the Group C, 2.20 ± 1.88 μg/L in the Group A and 1.54 ± 2.07 in the Group B. Statistical significant differences in mean concentration were found between the Group C and the Group A (*p* < 0.001) and between the Group C and the Group B (*p* < 0.001) (Table 2).


**Table 2.** Difference in the mean concentration of titanium, nickel vanadium and arsenic in the three groups.

\* The mean difference is significant at the 0.05 level. HSD, honestly significant difference.

#### **4. Discussion**

The aims of this study were to detect salivary metallic ions levels of Ti, V, Ni and As in patients with and without dental implants.

The metal concentration was measured by the use of ICP-MS, a versatile, rapid and extremely sensitive analytical technique used to determine different metallic and non-metallic inorganic substances present in concentrations lower than one part per billion.

Just a few studies have investigated the possible relationship of metallic ions dissolution and peri-implantitis, mainly focusing on titanium levels [27–29]

Safioti et al. first hypothesized a possible association between high titanium levels and implants affected by peri-implantitis, studying the metal concentration in submucosa plaque. The plaque samples were taken from the deepest points of the pockets of each implant through dental scalers (mini-five 1–2 Gracey) and then analyzed by ICP-MS. Results showed that implants with peri-implantitis had significantly (*p* < 0.05) higher titanium levels than healthy implants [27].

Olmedo et al. using the exfoliative cytology technique, measured the concentration of titanium particles in the peri-implant mucosa cells and found higher values in the group of patients affected by peri-implantitis compared to the group of patients with healthy implants [28,32].

In our study setting, attention was focused not only on titanium but also on other metals such as V, Ni and As.

Vanadium is ubiquitous: it is present in water and soil and its effects at systemic level are still debated: several authors have reported the use of vanadium in the treatment of diabetes mellitus, while others have correlated the long-term exposure with risk of cancer [33–36].

Nickel is known to be the cause of contact dermatitis and in general of allergic episodes, which can develop with serious consequences in the most sensitive subjects. Adverse reactions were documented in relation to orthodontic devices (brackets, arches) containing nickel. This element has a carcinogenic and mutagenic effects; therefore, exposure must be minimized [37–39].

Arsenic is one of the most widespread elements in nature: it can be found in soil, water, air and almost in all animal and plant tissues.

Arsenic poisoning can be acute (lethal) or chronic, caused by prolonged exposure even at low concentrations.

Martin-Camean et al. in 2014, in an in vivo study, evaluated the dissolution of different metal ions in the oral mucosa of subjects with orthodontic mini-implants. The epithelial cells of the oral mucosa were taken from each patient using a rubber brush. The samples were then analyzed through ICP-MS. The results reported only traces of vanadium, while the release of other elements occurred in the following growing order Cr < Ni < Ti < Cu < Al [40–42].

In our study, concentrations of metallic ions were higher in subjects with dental implants, either healthy or affected by peri-implantitis, compared to the control group, with the exception of vanadium.

Based on our results, metallic ions are detectable in the saliva of patients with dental implants; according to a recent systematic review conducted by Noronha Oliveira et al., degradation products—in the form of micro and nanoscale particles—can be found for different reasons, such as detachment from implant surface during surgical insertion, wear caused by micro-movements between contacting surfaces at implant connections, corrosive effects of therapeutic substances, like bleaching agents and fluorides or peri-implantitis treatment (mainly implantoplasty) [43].

In the present study, there were no statistically significant differences (*p* > 0.05) for metallic ions concentrations between patients with healthy dental implants and peri-implantitis subjects.

Our results are in contrast with the findings of Safioti et al. and Olmedo et al. [27,28]. However, in the above-mentioned articles, exclusion criteria did not limit enrollment to patients without other metallic prosthetic reconstructions and a control group without implants was not provided.

A recent systematic review by Gomes et al. [44] evaluated the diagnostic accuracy of biomarker levels in saliva to distinguish between healthy implants and implants affected by peri-implantitis. Based on their results, there was no clear evidence to support the use of salivary biomarkers (IL-6, IL-1β) in peri-implantitis detection. Pettersson et al. [45] performed gingival biopsies during surgical treatment of patients with severe periodontitis or peri-implantitis in order to evaluate titanium levels via ICP-MS. They found higher titanium values in peri-implantitis patients (*p* < 0.001) compared to periodontitis subjects, however samples were obtained after surgical treatment of dental implants and titanium release may be, therefore, exacerbated by ultrasonic scaling, as previously demonstrated by Eger et al. [46].

Furthermore, human samples from appropriate control groups (healthy implants and patients without metallic reconstructions) could not be obtained for ethical reasons.

Therefore, saliva collection and analysis through ICP-MS, as performed in the present study, seem a viable alternative to investigate data obtained by different populations.

#### **5. Conclusions**

Current evidence on the possible role of titanium or other metal particles in peri-implantitis pathogenesis is still controversial: in a recent critical review, Mombelli et al. concluded that there is poor specificity for the association of titanium particles and peri-implantitis, since metallic ions can be commonly detected in healthy and diseased peri-implant mucosa, as well as in gingiva of subjects without dental implants, being Tio2 used in multiple kinds of foods, toothpastes, cosmetics or medical pills [10].

Schwarz et al. in the latest World Workshop on the Classification of Periodontal and Peri-Implant Diseases and Conditions [18] stated that there was insufficient available evidence to consider titanium particles as a risk indicator for peri-implantitis.

Based on our results, concentrations of metallic ions were higher in subjects with dental implants, either healthy or affected by peri-implantitis, compared to the control group, with the exception of vanadium. No statistically significant differences were found in the metallic concentrations of healthy implants or peri-implantitis.

Future research should be orientated in conducting further studies, with a larger sample and a longitudinal design, to determine the role of titanium or other metals in peri-implantitis pathogenesis.

**Author Contributions:** P.P. and A.R. drafted the article, revised the paper and gave substantial contributions to the conception of the work; G.P. and A.D. gave substantial contributions to the conception of the work and revised the manuscript critically for important intellectual content. N.P. revised the paper, gave substantial contributions to the conception of the work and performed the statistical analysis. B.D.M. and P.C.P. revised the paper, gave substantial contributions to the conception of the work and recruited patients. M.B. gave substantial contributions to the conception of the work, revised the manuscript critically for important intellectual content and performed the ICP-MS analysis. All authors have read and agreed to the published version of the manuscript.

**Conflicts of Interest:** The authors declare they have no conflicts of interest related to this study.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article* **Salivary Biomarkers and Their Correlation with Pain and Stress in Patients with Burning Mouth Syndrome**

**Pia Lopez-Jornet 1,\*, Candela Castillo Felipe 2, Luis Pardo-Marin 3, Jose J. Ceron 3, Eduardo Pons-Fuster <sup>4</sup> and Asta Tvarijonaviciute <sup>3</sup>**


Received: 22 February 2020; Accepted: 25 March 2020; Published: 28 March 2020

**Abstract:** Objective: To evaluate a panel of salivary analytes involving biomarkers of inflammation, stress, immune system and antioxidant status in patients with burning mouth syndrome (BMS) and to study their relationship with clinical variables. Materials and Methods: A total of 51 patients with BMS and 31 controls were consecutively enrolled in the study, with the recording of oral habits, the severity of pain using a visual analogue scale (VAS), the Hospital Anxiety and Depression (HAD) score and the Oral Health Impact Profile-14 (OHIP14) score. Resting whole saliva was collected with the drainage technique, followed by the measurement of 11 biomarkers. Results: The salivary flow was higher in patients with BMS. Among all the biomarkers studied, significantly higher levels of alpha-amylase, immunoglobulin A (IgA), and macrophage inflammatory protein-4 (MIP4) and lower levels of uric acid and ferric reducing activity of plasma (FRAP) were observed in the saliva of patients with BMS as compared to the controls (*p* < 0.05 in all cases). Positive correlations were found between pain, oral quality of life and anxiety scores and salivary biomarkers. Conclusions: BMS is associated with changes in salivary biomarkers of inflammation, oxidative stress and stress, being related to the degree of pain and anxiety.

**Keywords:** saliva; IgA; alpha-amylase; uric acid; stress; inflammation

#### **1. Introduction**

The International Association for the Study of Pain (IASP) defines burning mouth syndrome (BMS) as an intraoral burning or dysesthetic sensation that manifests daily for more than two hours over three months, with no evidence of clinical lesions [1]. The epidemiological data on BMS are largely contradictory, in part because of a lack of strict compliance with the diagnostic criteria of the disorder. Nevertheless, BMS is estimated to affect 4% of the general population and 18–33% of all postmenopausal women [2–6]. The symptoms are generally focused on the tongue and lips and are almost always bilateral. The palate and other locations within the oral cavity are less commonly affected. Although much has been published about BMS, its underlying etiopathogenesis is largely unknown, and complex interactions among local, systemic and psychogenic factors are believed to be involved [2,7,8]. For these reasons, BMS is poorly diagnosed, and in many cases management is deficient, a situation that causes frustration for both physicians and patients [3,5]. In the same way as with other chronic pain syndromes, BMS is characterized by associated psychological problems. Changes in personality and mood state, particularly anxiety and depression, are a typical finding in patients with BMS [2,9].

The use of saliva as a sample for clinical analysis has increased in recent decades. Saliva is a body fluid that reflects the physiological condition of the body. It can be easily collected using non-invasive and relatively inexpensive methods [4,5], and it is gaining attention as a source of biomarkers in different oral and systemic pathologies [10]. In BMS an increase in amylase (a marker of the adrenergic system) and immunoglobulin A (IgA) (a marker of the immune system) have been found [11–14]. However, to the authors' knowledge, there have been no studies in which a panel of analytes reflecting possible pathogenic and psychological factors were evaluated in this disease, and also compared with the degree of stress and pain.

The objective of this study was to evaluate how a panel of biomarkers of inflammation represented by complement C4 (CC4), α1-antitrypsin (a1AT), C-reactive protein (CRP), macrophage inflammatory protein-4 (MIP4), pigment epithelium-derived factor (PEDF), serum amyloid P (SAP), haptoglobin (Hp), a panel of biomarkers of oxidative stress integrated by uric acid and ferric reducing activity of plasma (FRAP), the salivary alpha-amylase (sAA) as a marker of the adrenergic system and total immunoglobulin A (IgA) as a marker of the innate immune system can be altered in patients with BMS, and study their possible relation with the degree of pain and stress of the patients was measured by visual analogue scale (VAS) and the Hospital Anxiety and Depression (HAD) score. Also, the possible influence of oral health was evaluated.

#### **2. Material and Methods**

#### *2.1. Study Design and Subjects*

A cross-sectional study was carried out with data compiled between January 2018 and September 2019 in the Dental Clinic of the University of Murcia (Murcia, Spain). All the patients were informed about the study and gave consent to participation in the trial, which was approved by the local Clinical Research Ethics Committee (Ref.: 2203/2018). The following inclusion criteria were established: patients over 18 years of age with a diagnosis of BMS based on exhaustive oral examination, laboratory findings and the patient profile and symptoms [15], presenting continuous symptoms of oral burning or pain persisting for at least two hours per day, lasting longer than three months, without paroxysms and not following any unilateral nerve trajectory, and with no clinical mucosal alterations. The patients underwent blood testing to confirm the absence of alterations (including blood count, glucose, serum iron, ferritin and transferrin, folic acid and vitamin B12 levels). Cancer patients and individuals with known liver or kidney disease were excluded, as were those with oral diseases other than BMS such as Sjögren's syndrome or ongoing infection, thyroid diseases, coagulation disorders, psychiatric disease, and pregnant or nursing women.

The healthy controls were recruited from the Dental School of the University of Murcia and consisted of patients with the same sociosanitary characteristics and matched for age and gender. Following the procurement of written informed consent, a structured questionnaire was administered to confirm the absence of significant medical conditions. The study was carried out following the STROBE (Strengthening the Reporting of Observational studies in Epidemiology) guidelines.

In all cases, patients were not included in the study if their samples showed hemolysis as determined by visual inspection [16,17].

#### *2.2. Data Collection*

A structured interview was used to collect sociodemographic and clinical information as well as data referring to smoking and alcohol consumption. The patients were evaluated by a trained professional (C.C.F.) who conducted the interviews and explorations, administered the questionnaires and collected the saliva samples.

An extraoral and intraoral exploration was carried out, documenting the presence of caries and missing teeth. Pain intensity was scored using a visual analogue scale (VAS) (0 = no pain and 10 = worst possible pain) [18].

The Hospital Anxiety and Depression (HAD) scale [19] was used to assess the emotional state of the participants. This instrument consists of two subscales relating to anxiety (HAD-A) and depression (HAD-D). Concerning the interpretation of the HAD scale, scores of > 10 indicate the probable presence of anxiety or depression, scores of ≤ 7 indicate no significant anxiety or depression, and scores of 8–10 are of borderline significance.

The evaluation of the oral quality of life, in turn, was based on the Oral Health Impact Profile-14 (OHIP14) score [20].

#### *2.3. Saliva Collection*

Resting whole saliva was collected using a standardized method [18]. To avoid possible contamination from other sources, the patients were instructed to rinse the mouth thoroughly before saliva sample collection. The subjects were required to avoid heavy physical exercise one hour before sampling. Unstimulated saliva was obtained using the draining method for 5 min. The samples were collected at about the same time in all subjects (8:00 to 11:00 a.m.). The saliva was vortexed and centrifuged (3000× *g* for 10 min at 4 ◦C) immediately after collection, and the supernatant was transferred into polypropylene tubes and stored at −80 ◦C until analysis.

#### *2.4. Biochemical Analysis*

Complement C4, a1AT, CRP, MIP4, PEDF and SAP in saliva were analyzed using a commercially available kit (Human Neurodegenerative Disease Magnetic Bead Panel 2, Neuroscience Multiplex Assay; Life Science, Darmstadt, Germany) according to the manufacturer's instructions. Values were calculated based on a standard curve constructed for the assay. Saliva total protein quantification was done using a commercially available colorimetric kit for measuring urine and low-complexity region (LCR) proteins (protein in urine and CSF, Spinreact, Barcelona, Spain) and validated for human saliva [21]. Uric acid was measured using a colorimetric commercial kit (Uric acid, Beckman Coulter Inc., Fullerton, CA, USA) following the International Federation of Clinical Chemistry and Laboratory Medicine (IFCC) method [21]. Total IgA was evaluated with a commercial ELISA kit (Bethyl, Montgomery, TX, USA) previously validated for use with human saliva samples [18]. FRAP (ferric reducing ability of plasma) measurement was based on the method described by Benzie and Strain [22] with some modifications [23]. Hp levels were measured in saliva using a homemade immunoassay as previously described [24]. Salivary alpha-amylase (sAA) activity was measured using a colorimetric commercial kit (Alpha-Amylase, Beckman Coulter Inc., Fullerton, CA, USA) following the IFCC method as previously reported and validated [25].

#### *2.5. Statistical Analysis*

For the descriptive statistical analysis of the sample, the basic descriptive methods were used, with calculation of the frequencies, mean, median, standard deviation (SD) and 25th and 75th percentile values. The comparison of means between groups was based on the Student's t-test or Mann–Whitney test, depending on the data distribution as verified with the Kolmogorov–Smirnov test, and the homogeneity of groups was confirmed. Clinical scores, smoking and alcohol consumption habits, as well as sex distribution among the two groups, were explored using the chi-square test. The correlations between variables were checked using the partial correlation corrected by age and sex. Previous studies that also analyzed biomarkers in saliva in patients with burning mouth syndrome, such as sAA and total IgA, have been able to detect differences with 30 or fewer individuals in each group considering α = 0.05 and β = 0.20 [11]. Based on these data we assumed that the study was sufficiently powered (ncontrols = 31; nBMS = 51) to achieve our aims. Differences and associations among groups were considered statistically significant when *p* < 0.05.

#### **3. Results**

The study sample consisted of 82 consecutively enrolled individuals (71 females and 11 males), of which 51 were diagnosed with BMS while the remaining 31 constituted the control group. Table 1 describes the characteristics of each group. Statistically significant differences were not detected in sex, age or smoking and alcohol consumption habits between the two groups of patients (*p* > 0.05).


**Table 1.** General characteristics of participants.


**Table 1.** *Cont.*

a, Student's t-test; b, Mann–Whitney test; c, Pearson's chi-square. Bold type denotes statistical significance.

Concerning oral hygiene and health, the controls and patients with BMS did not present statistically significant differences in habits related to tooth brushing and use of mouthwashes, while patients with BMS used more dental floss than controls (*p* < 0.01). In turn, controls presented a comparatively greater number of caries, though the difference was not statistically significant (*p* > 0.05), and significantly fewer missing teeth compared with the BMS group (*p* < 0.01).

Of the clinical variables, the mean burning sensation score in patients with BMS was greater than 8, while it was equal to 0 in healthy controls. The OHIP14 score was 2.3-fold higher in the BMS group as compared with the control group (*p* = 0.001).

Concerning the psychological profile as evaluated by the HAD scale, the anxiety scores among the patients were 2.3-fold higher than in controls (*p* < 0.001). While HAD-D, although being 1.4-fold higher in patients, did not show statistically significant differences between the two groups (*p* > 0.05).

Resting whole saliva flow was 1.4-fold greater in the BMS group as compared to healthy controls (*p* < 0.05) (Figure 1a and Table S1). The study of biomarkers in resting whole saliva yielded significant differences between the cases (patients with BMS) and controls for sAA (*p* < 0.01) and IgA (*p* < 0.05) (Figure 1a,b and Table S1).

**Figure 1.** Salivary flow rate (**a**), salivary alpha-amylase (sAA; (**b**)) and immunoglobulin A (IgA; (**c**)) in healthy controls (*n* = 31) and patients with burning mouth syndrome (BMS; *n* = 51). Differences of medians and means±SD for salivary flow are 0.54 and 0.75 ± 0.41, for sAA they are 48,660 and 109,536 ± 41,492 and for IgA they are 22.47 and 34.01 ± 12.72.

When absolute values of salivary biomarkers were corrected by salivary flow, sAA, IgA and MIP4 showed statistically higher concentrations in patients with BMS as compared with healthy controls (Figure 2 and Table S1). When values were corrected by total protein content, statistically significantly higher levels in patients with BMS were observed for sAA and lower for uric acid and FRAP (Figure 3 and Table S1).

**Figure 2.** Salivary alpha-amylase (sAA; (**a**)), immunoglobulin A (IgA; (**b**)) and macrophage inflammatory protein-4 (MIP4; (**c**)) corrected by salivary flow in healthy controls (*n* = 31) and patients with burning mouth syndrome (BMS; *n* = 51). Difference of medians and means ± SD for sAA are 44.0 and 47.4 ± 15.91, for IgA are 119.9 and 147.6 ± 38.4 and for MIP4 are 13.0 and 33.6 ± 27.2.

**Figure 3.** Salivary alpha amylase [sAA; (**a**)], uric acid (**b**) and ferric reducing activity [FRAP; (**c**)] corrected by salivary total protein content in healthy controls (*n* = 31) and patients with burning mouth syndrome (BMS; *n* = 51). Difference of medians and means ± SD for sAA are 38,525 and 33,653 ± 15,365, for uric acid are –11.5 and –7585 ± 4593 and for FRAP are –0.17 and –0.1413 ± 0.06614.

Partial correlation data are given in Table S2. When absolute values were evaluated, a positive correlation was detected between VAS and total proteins (*r* = 0.241; *p* < 0.05), sAA (*r* = 0.260; *p* < 0.05) and IgA (*r* = 0.334; *p* < 0.01); HAD-A was correlated with total proteins (*r* = 0.251; *p* < 0.05), IgA (*r* = 0.302; *p* < 0.05), PEDF (*r* = 0.351; *p* < 0.01) and MIP4 (*r* = 0.267; *p* < 0.05) and HAD-D with IgA (*r* = 0.261; *p* < 0.05), PEDF (*r* = 0.363; *p* < 0.01) and MIP4 (*r* = 0.271; *p* < 0.05); OHIP14 was correlated with IgA (*r* = 0.273; *p* < 0.05).

After data were corrected by salivary flow, statistically significant correlations remained between VAS and sAA (*r* = 0.269; *p* < 0.05) and IgA (*r* = 0.367; *p* < 0.01); between HAD-A and IgA (*r* = 0.338; *p* < 0.01), PEDF (*r* = 0.297; *p* < 0.05) and MIP4 (*r* = 0.264; *p* < 0.05); between HAD-D and PEDF (*r* = 0.386; *p* = 0.01) and MIP4 (*r* = 0.269; *p* < 0.05) and between OHIP14 and IgA (*r* = 0.313; *p* < 0.01).

After data were corrected by salivary flow, negative correlations appeared between sAA and OHIP14 (*r* = −0.242; *p* < 0.05) and between HAD-A and FRAP (*r* = 0.250; *p* < 0.05) and CRP (*r* = −0.243; *p* < 0.05).

#### **4. Discussion**

Burning mouth syndrome is a clinically relevant form of chronic orofacial pain [1–3]. The diagnosis of the syndrome remains a challenge for health professionals due to the discrepancy between the intensity of pain as reported by the patient and the absence of objective clinical lesions. In this study we therefore focused on the identification of potential biomarkers capable of reflecting the physiopathological changes involving pain, psychological stress and inflammation that occur in the context of the disease. Of 11 biomarkers evaluated in the present study, three—sAA, IgA and MIP4—were found to be increased, and two—uric acid and FRAP—were found to be decreased in the saliva of patients with BMS versus controls, suggesting that these patients can present with alterations in immune response, pro-inflammatory status and oxidative stress that are related to pain sensation and psychological stress.

In the present study, a higher salivary flow rate was detected in patients with BMS as compared with healthy controls. In the scientific literature, controversy exists regarding this topic since some authors did not detect changes in salivary flow between controls and patients with BMS [12–14], while others observed lower flow rates in patients with BMS [9,11,26]. Similarly, different authors reported divergent results related to total protein content in the saliva of patients with BMS, as some of them detected decreased [27], increased [12,14] or unchanged (present study) salivary total protein content in these patients when compared with controls. The characteristics of the method or device used to collect the samples and the analytical assays used may have influenced the results obtained and, therefore, the differences among the different studies. It is therefore important to follow the same guidelines for sample collection and processing to enable comparison of the results from different studies.

Since components present in saliva have different origins, i.e., are locally secreted by salivary glands or oral mucosa, or pass from blood among other, and can be affected by non-uniform salivary flow rate, the need to correct values of salivary biomarker by flow rate, total protein content or salivary osmolarity versus use of their absolute value is being increasingly acknowledged and studied by different authors [25,28]. In this study we aimed to evaluate the results without any adjustment, and we also corrected salivary flow or total protein content to gain knowledge about the possible effect that these corrections can have in this particular disease. When levels of biomarkers without any adjustment by saliva flow or protein content were compared between healthy and diseased patients, statistically significant changes were detected only in sAA and IgA. Correction by salivary flow rate allowed identification of the increased levels of MIP4 in the saliva of patients with BMS but also yielded higher differences between the studied groups for IgA in accordance with previously reported data [29]. When values were corrected by total protein content the presence of oxidative stress in patients with BMS was detected, as levels of uric acid and FRAP were lower compared to controls. Therefore, in BMS, correcting salivary values by flow rate and total protein content would be recommended in this disease since it can detect additional changes in salivary analytes that are correlated with the clinical condition of the patient.

sAA is considered to be a sensitive biomarker of stress-related changes that reflect the activity of the sympathetic nervous system (SNS) [30]. Most patients with BMS experience chronic pain and have a poorer quality of life than do healthy individuals [9]. Furthermore, chronic pain characterizing BMS was associated with psychological problems, specifically personality and mood changes, anxiety and depression. The results of this study agree with these findings and previous reports that detected higher sAA levels in the saliva of patients with BMS [12,14] since patients included in the present study showed both higher anxiety scores and sAA levels; however, another study did not detect changes in sAA between healthy controls and patients with BMS, and differences in the severity of the disease in the population evaluated or in the assay used could be the cause of these differences [9]. A weak positive correlation between sAA (expressed in absolute concentrations and when corrected by flow) and burning sensation was detected, which could indicate that increases in this enzyme could be influenced in part by pain.

Salivary IgA was another biomarker that showed statistically significant differences among controls and patients in the studied population, being higher in patients with BMS. These data are consistent with those reported by other authors [12–14]. Furthermore, a higher difference between the two groups was detected when IgA was corrected by salivary flow resulting in an increase of 1.8-fold to 2.9-fold change between the two groups. IgA is an immune glycoprotein that acts as a defense against pathogens [31]. Nevertheless, different studies have suggested that changes in salivary IgA can also be associated with stress [32]. This hypothesis would be supported by the observed positive correlations between IgA in saliva and oral quality of life, burning sensation and anxiety in the studied population. In addition, MIP, a biomarker of the innate immune system [33], was also increased in patients with BMS and weakly positively correlated with anxiety and depression scores, suggesting that the immune system is implicated in the pathogenesis of BMS.

In the present study, when the results were corrected by protein content, patients with BMS had lower salivary FRAP and uric acid, suggesting the presence of oxidative stress, and in the case of FRAP, it was weakly correlated with HAD-A, which would indicate that this increase could be influenced by stress. The two markers are closely related since uric acid was the main component (up to 60%) of the FRAP assay [22]. Uric acid is considered to be the most important antioxidant molecule in saliva since it is responsible for approximately 70% of the total antioxidant capacity of this body fluid. As a reactive oxygen species scavenger, it helps stabilize arterial pressure and oxidative stress [21,34–36]. In turn, alterations in its levels in saliva were associated with acute stress [34] and different local and systemic pathologies such as oral lichen planus [37] or nephropathies [38]. Previous studies that evaluated salivary uric acid in patients with BMS did not find statistically significant changes as compared with healthy controls [16,20], although these studies did not make corrections for total protein content.

A limitation of the study is the fact that the measurements were limited to a single time-point. Variations of these markers in the same individual should be investigated in the context of future studies with a larger sample size to increase the power, and by employing multiple samplings over time, resulting in more accurate estimations. Furthermore, it would be of interest to apply the biomarkers that showed significant changes in a larger population to corroborate the results of this study. Finally, the possible blood contamination in saliva samples should have been evaluated using more sensitive methods, such as determination of hemoglobin or transferrin, although these methods were shown to be affected by different factors such as age, hormones and salivary flow among others [16]. Conversely, visual inspection was shown to be sufficient in the case of determining some of the analytes including oxidative stress markers in saliva without interfering with the results [17].

#### **5. Conclusions**

In conclusion, patients with BMS showed changes in biomarkers associated with stress such as sAA and IgA, with the immune system such as MIP4 and oxidative stress such as uric acid and FRAP in saliva as compared with healthy controls, which are related to clinical variables including burning sensation and anxiety. Moreover, in this particular disease, the study of the absolute values but also the values corrected by flow and total protein would be recommended. Overall, biomarkers that were shown to change differently between groups could potentially help clinicians not only with diagnosis but also in objectively evaluating the severity of this disease, although further large-scale studies should first be performed to collaborate our findings.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2077-0383/9/4/929/s1: Table S1: Median and interquartile range data of salivary markers in healthy controls (n = 31) and patients suffering from BMS (n = 51); Table S2: Partial correlation adjusted for age and sex.

**Author Contributions:** Conceptualization, P.L.-J., C.C.F. and A.T.; Methodology, P.L.-J., C.C.F. and A.T.; Software, L.P.-M., J.J.C. and A.T.; Validation, P.L.-J., C.C.F. and E.P.-F.; Formal Analysis, P.L.-J., A.T., L.P.-M. and J.J.C.; Investigation, P.L.-J., C.C.F., L.P.-M. and A.T.; Resources, P.L.-J., C.C.F., J.J.C. and A.T.; Data Curation, P.L.-J. and C.C.F.; Writing-Original, P.L.-J., J.J.C. and A.T.; Draft Preparation, E.P.-F.; Writing-Review & Editing, P.L.-J., A.T. and J.J.C.; Supervision, All authors; Project Administration, P.L.-J. and A.T. All authors have read and agreed to the published version of the manuscript.

**Funding:** A.T. has a post-doctoral fellowshisp "Ramón y Cajal" supported by the "Ministerio de Economía y Competitividad", Spain. This work was supported by a grant from the Program for Research Groups of Excellence of the Seneca Foundation, Murcia, Spain (grant 19894/GERM/15).

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article* **Enhanced Inflammation and Nitrosative Stress in the Saliva and Plasma of Patients with Plaque Psoriasis**

#### **Anna Skutnik-Radziszewska 1, Mateusz Maciejczyk 2, Iwona Flisiak 3, Julita Krahel 3, Urszula Kołodziej 4, Anna Kotowska-Rodziewicz 4, Anna Klimiuk <sup>5</sup> and Anna Zalewska 5,\***


Received: 10 February 2020; Accepted: 9 March 2020; Published: 10 March 2020

**Abstract:** Psoriasis is the most common inflammatory skin disease, characterized by the release of proinflammatory cytokines from lymphocytes, keratinocytes, and dendritic cells. Although psoriasis is considered an immune-mediated inflammatory disease, its effect on secretory activity of salivary glands and quantitative composition of saliva is still unknown. The aim of this study was to evaluate the secretion of saliva as well as several selected inflammation and nitrosative stress biomarkers in unstimulated and stimulated saliva as well as plasma of psoriasis patients. We demonstrated that, with progressing severity and duration of the disease, the secretory function of the parotid and submandibular salivary glands is lost, which is manifested as decreased unstimulated and stimulated saliva secretion and reduced salivary amylase activity and total protein concentration. The levels of tumor necrosis factor-alpha (TNF-α), interleukin-2 (IL-2), and interferon-gamma (INF-γ) were significantly higher, whereas interleukin-10 (IL-10) content was considerably lower in unstimulated and stimulated saliva of patients with psoriasis compared to the controls, and the changes increased with the disease duration. Similarly, we observed that the intensity of nitrosative stress in the salivary glands of psoriasis patients depended on the duration of the disease. By means of receiver operating characteristic (ROC) analysis, we showed that the evaluation of nitric oxide (NO), nitrotyrosine, and IL-2 concentration in non-stimulated saliva with high sensitivity and specificity differentiated psoriasis patients on the basis of the rate of saliva secretion (normal salivation vs. hyposalivation). In summary, the dysfunction of salivary glands in psoriasis patients is caused by inflammation and nitrosative stress.

**Keywords:** plaque psoriasis; salivary glands; saliva; cytokines; nitrosative stress

#### **1. Introduction**

Psoriasis vulgaris is a skin inflammatory disease, the third most common among autoimmune diseases [1]. The pathogenesis of psoriasis involves the combination of genetic susceptibility, aberrant immune response, and several environmental factors (injuries, viral infections, medications taken, food intolerances). Typical features of psoriasis vulgaris include an immune-mediated process in which the key role is played by Th1 cells (T-helper 1). The presence of antigen-specific CD4+ (dendritic

cell) T cells secreting type 1 cytokines: interferon-gamma (INF-γ), interleukin-2 (IL-2), and tumor necrosis factor-alpha (TNF-α) was observed in psoriatic skin lesions [2,3]. Moreover, the imbalance between Th1 and Th2 cells in psoriasis was confirmed by studies that showed interleukin-10 (IL-10) deficiency in psoriatic skin lesions [4]. Recent findings suggest that redox imbalance in the blood and skin of patients with psoriasis, resulting from the immune system stimulation, plays as important role in the pathogenesis of plaque psoriasis as the inflammatory process itself. It has been shown that, in the course of psoriasis, oxidative stress (OS) is primarily caused by reduced activity/concentration of antioxidants [5–9] and leads to increased oxidative modification of cellular elements of skin and plasma [6,10,11]. Plasma and erythrocyte product of lipid peroxidation (malondialdehyde, MDA) is considered a biomarker of plaque psoriasis exacerbation [12,13].

The detrimental inflammatory milieu and increased production of free radicals (ROS) associated with plaque psoriasis are not limited to the skin, but are also responsible for the growing number of comorbidities, including cardiological diseases, metabolic syndrome, chronic kidney disease, mood disorders, and salivary gland diseases [5,14–17].

Saliva produced by salivary glands performs numerous important functions in the oral cavity: hydrating it, removing harmful waste products and bacteria, participating in the remineralization of dental hard tissues, maintaining redox balance, and being involved in immune responses [18– 20]. Disorders of both the composition and amount of saliva secreted into the oral cavity have physiological and psychological consequences. Therefore, it is very important to understand the mechanisms leading to salivary gland dysfunction in the course of systemic diseases. Unfortunately, the pathophysiology of salivary gland disorders in psoriasis is still unknown. In our previous work, we demonstrated that plaque psoriasis is accompanied by salivary redox imbalances with the prevalence of oxidation reactions. We observed that redox equilibrium in the submandibular glands was more vulnerable, and antioxidant capacity of the submandibular glands decreased with the disease duration [15]. There have been very few studies on the modification of saliva inflammatory components in psoriatic patients. Ganzetti et al. [21] demonstrated higher levels of TNF-α, transforming growth factor-beta (TGF-β), and interleukin-1 (IL-1β) in the saliva of psoriatic subjects vs. the controls. Unfortunately, most patients with psoriasis had also been diagnosed with periodontitis or gingivitis. Consequently, the observed cytokine changes in saliva reflected periodontal inflammation and not psoriasis-related salivary changes, and thus they did not explain the pathophysiology of salivary gland dysfunctions in the course of this disease. It has been evidenced that pro-inflammatory cytokines boost the expression of the inducible nitric oxide (NO) synthase (*i*NOS) in the cells, which results in increased NO synthesis in inflamed joints [22]. It was proven that NO and other reactive nitrogen intermediates affect a vast number of physiological functions of salivary glands, including exocytosis, water secretion, salivary blood flow, and non-specific immunological reactions [23–25]. The contribution of nitrosative stress to the pathophysiology of salivary glands in the course of psoriasis is unknown.

An ideal biomarker is characterized by simple determination as well as high sensitivity, specificity, and repeatability. It should also enable the identification of the patient's physiological and pathological status and response to the applied treatment [26]. Numerous biomarkers have been suggested for easier diagnosis of psoriasis and monitoring of its treatment. However, studies on psoriasis biomarkers have been based on blood tests, examining skin fragments, conducting genetic tests, and transcriptomics. The results of these studies were divergent, and therefore neither was considered reliable nor was accepted as a psoriasis marker [27]. Saliva is a mixture of secretions of large salivary glands and gingival fluid. It also contains almost all the elements present in the blood and passing through the spaces between cells as part of an inter- and paracellular transport. It is also relatively easy and safe to collect, providing a new, non-invasive way to diagnose numerous general diseases [28–32].

Thus, the aim of this work was to explore the mechanisms responsible for salivary gland dysfunction in psoriasis. We compared the concentrations of TNF-α, IL-2, INF-γ, IL-10, NO, peroxynitrite, S-nitrosothiols, and nitrotyrosine in the saliva and blood of psoriatic subjects with hyposalivation and normal salivation vs. healthy controls. The goal of our study was also to search for salivary biomarkers to assess the severity of psoriasis and the accompanying salivary complications.

#### **2. Materials and Methods**

We obtained the consent of the Local Research Ethics Committee in Bialystok (permission number: R-I-002/563/2018). All patients as well as healthy subjects were informed about the purpose of the study and its risks and benefits, and they all consented to the collection of saliva and blood samples.

This experiment included psoriatic patients applying for treatment to the Department of Dermatology and Venereology of the Medical University of Bialystok. The reason for reporting to the hospital was exacerbation of psoriasis symptoms. Patients had not undergone general treatment of psoriasis within 2 preceding years. The only acceptable forms of treatment included local application of ointment with glucocorticosteroids, but not during 3 preceding months. The severity of skin lesions was assessed using the previously described Psoriasis Area and Severity Index (PASI) [15].

A total of 60 healthy patients participated in the study, who were individuals reporting for check-up visits to the Department of Restorative Dentistry at the Medical University of Bialystok, and were matched to the group of patients in terms of age and sex.

Within 6 preceding months, participants did not take any medicines that could affect the composition and secretion of saliva, as well as vitamins, antioxidants, and antibiotics. The participants from the control group were generally healthy, and patients with psoriasis did not have any accompanying diseases. The inclusion criteria was absence of periodontal pathology (periodontal pocket depth (PPD) < 2 mm, did not bleed during the probing) and no inflammatory or fungal changes on the oral mucosa. Only subjects without acrylic dentures were included in the study. The patients as well as healthy controls were not addicted to alcohol and did not smoke cigarettes.

One of the criteria differentiating the dysfunction of salivary glands is reduced flow of unstimulated saliva (NWS) [18]. Therefore, to delineate the influence of psoriasis vulgaris on the parotid and submandibular glands, psoriatic patients were divided into two groups: with reduced NWS (psoriasis hyposalivation-PH, *n* = 30) and normal NWS (psoriasis normal secretion- PN, *n* = 30). The limit value for NWS was assumed at ≤ 0.2 mL/min, which is considered the minimum value of NWS for healthy population [18]. Moreover, the minimum value of NWS in our control group was 0.21 mL/min.

The number of patients was confirmed by the test as sufficient, and the test power was 0.9. Clinical characteristics of the patients are presented in Table 1.


**Table 1.** Clinical characteristics of plaque psoriasis patients and control subjects.


**Table 1.** *Cont.*

Abbreviations: ALT—alanine transferase; AST—aspartate transaminase; CRP—C-reactive protein; DMFT—decayed, missing, filled teeth index; GI—gingival index; Glc—D-glucose; HCT—hematocrit; *n*—number of patients; PASI—Psoriasis Area and Severity Index; PBI—papilla bleeding index; PLT—platelets; RBC—red blood cells; WBC—white blood cells. NDnot defined.

#### *2.1. Blood Collection*

Blood was collected at fasting, either at the admission of a patient to the hospital or during routine tests in case of control subjects. Blood was collected at 5 mL using an S-Monovette EDTA K3 tube (Sarstedt, Nümbrecht, Germany). The samples were then centrifuged (3000 × *g*, 10 min, 4 ◦C). No hemolysis was observed in any of the obtained plasma samples. To prevent sample oxidation, 0.5 M Butylated hydroxytoluene BHT (Sigma-Aldrich, Saint Louis, MO, USA; 10 μL/mL blood) was added [28]. Plasma was frozen (−82 ◦C). The samples were stored deep-frozen for no longer than 6 months.

#### *2.2. Saliva Collection*

The examined material was unstimulated and stimulated (SWS) total saliva collected from the patient by the spitting method [28]. Saliva was collected in the morning, on an empty stomach, between 8 a.m. and 10 a.m. in order to minimize the effect of daily changes on saliva secretion. The participants had refrained from taking any drugs for 8 hours prior to the examination. Saliva was collected in a separate room, in a sitting position with the head slightly inclined downwards, with minimized face and lip movements, and after a 5-minute adaptation period. Next, the patient rinsed the mouth three times with water at room temperature. The saliva collected during the first minute was discarded. The subsequent batches of saliva (the patient actively spat out the saliva accumulated in the bottom of the oral cavity) were collected into a plastic centrifuge tube placed in an ice container. NWS was collected for 15 minutes, and SWS was collected after a 5-minute break. Stimulation was performed by dropping 100 μL of 2% citric acid on the tip of the tongue every 20 seconds, for 5 minutes. To prevent sample oxidation, 0.5 M BHT (Sigma-Aldrich, Saint Louis, MO, USA; 10 μL/mL blood) was added to the saliva [33]. The volume of each sample was measured with a pipette calibrated to 0.1 mL. Saliva secretion was calculated by dividing the volume of the obtained saliva by the number of minutes of its collection. The collected saliva was centrifuged (20 minutes, 4 ◦C, 10,000 × *g*) [33]. The sediments

were discarded, and supernatant fluid was divided into portions of 200 μL each, frozen in −80 ◦C, and stored until assayed, but for no longer than 6 months.

#### *2.3. Stomatological Examination*

After saliva collection, stomatological examinations were performed by one dentist (A.S.-R.), in artificial light, using a dental mirror, probe, and periodontal probe (Hossa International, Warsaw, Poland; design and construction in accordance with WHO guidelines) according to WHO criteria [34]. The dental status of each participant was assessed on the basis of the DMFT index (decayed, missing, filled teeth). The condition of periodontal tissues was assessed using GI (gingival index) and probing pocket depth (PPD) were assessed at teeth 16, 21, 24, 36, 41, and 44. The PPD and occurrence of bleeding were assessed after gently introducing the probe into the gingival space parallel to the long axis of the tooth, to the depth of perceived resistance posed by the bottom of the gingival gap. Four surfaces of each examined tooth (mesial, distal, buccal, and lingual) were examined. In the case of GI, we used a scale of 0 to 3. The sum of the values, from four surfaces divided by 4, determined the level of the PPD/ gingival index for a given tooth. Then, the results were added and divided by the number of teeth examined. In 25 participants, the inter-rater agreement between the examiner (A.S.-R.) and other experienced dentists (U.K., A.K.) was performed. The reliability for DMFT was *r* = 0.99, for GI was *r* = 0.89, and for PPD was *r* = 0.94. If bleeding during probing or PPD deeper than 2 mm were found, previously collected saliva was discarded and the patient was excluded from the study. Thus, 53 patients with psoriasis and 37 healthy controls who had bleeding of probing and/or PPD > 2 mm were disqualified from the study.

#### *2.4. Biochemical Assays*

All determinations of plasma, NWS, and SWS were performed in duplicate samples. The absorbance/fluorescence was measured using an Infinite M200 PRO Multimode Microplate Reader (Tecan Group Ltd., Männedorf, Switzerland). The results were standardized to 1 mg of protein.

#### *2.5. Pro-Inflammatory Cytokines*

The concentrations of TNF-α, IL-2, IL-10, and INF-γ were determined by the ELISA method using commercial kits from EIAab Science Inc. Wuhan (Wuhan, China), according to the manufacturer's instructions.

#### *2.6. Nitrosative Stress*

The concentration of NO was determined spectrophotometrically using the Griess reagent—a solution of sulfanilic acid and α-naphthylamine in acetic acid [35]. The reaction of nitrates with sulfanilamide produces *N*-(1-naphthyl)-ethylenediamine dihydrochloride, the absorbance of which was measured at 490 nm wavelength.

The concentration of S-nitrosothiols was determined spectrophotometrically on the basis of the reaction of the Griess reagent with Cu2<sup>+</sup> copper ions [36]. The solution was shaken and set aside for 20 minutes, and then the absorbance was measured at 490 nm [37].

Peroxynitrite concentration was determined spectrophotometrically via a test using peroxynitrite decomposition followed by nitration of 4-hydroxyphenylacetic acid (4-HPA) and glycyltyrosine [38]. The reaction resulted in the production of nitrophenol, the absorbance of which was measured at 320 nm wavelength.

The concentration of nitrotyrosine was determined by ELISA using the Nitrotyrosine ELISA kit from Immunodiagnostik AG (Bensheim, Germany), following the manufacturer's instructions provided in the package.

#### *2.7. Salivary Protein*

Salivary protein content was determined using the bicinchoninic acid (BCA) method (Pierce BCA Protein Assay; Thermo Scientific (Rockford, IL, USA)). Bovine serum albumin was used as a standard.

#### *2.8. Salivary Amylase*

The activity of salivary amylase (EC 3.2.1.1.) was determined colorimetrically using 3,5-dinitrosalicylic acid (DNS) as a substrate [39]. By this method, starch was transformed by amylase to maltose, and was measured at 540 nm by the complex with DNS.

#### *2.9. Statistical Analysis*

The obtained results were assessed statistically by means of one-way analysis of variance (ANOVA). The significance of differences between individual groups was determined with the post-hoc Tukey's HSD test, and normal distribution was confirmed via the Shapiro–Wilk test. The correlations between the examined parameters were described using the Pearson correlation coefficient. The value of *p* < 0.05 was considered statistically significant. In order to assess the diagnostic usefulness between plaque psoriasis patients with normal salivary secretion and hyposalivation, receiver operating characteristic (ROC) curves were generated and then the area under the curve (AUC) was calculated. Every parameter had its optimal limit values determined, which simultaneously provided high sensitivity and specificity. The analysis of the data was performed in the statistical program GraphPad Prism 8.3.0 for MacOS.

#### **3. Results**

#### *3.1. Inflammatory Cytokines*

#### 3.1.1. NWS

The concentration of IL-2 (↑10.91%, *p* = 0.007; ↑33.64%, *p* < 0.0001, respectively) and INF- γ (↑33.11%, *p* ≤ 0.0001; ↑57.34%, *p* ≤ 0.0001, respectively) in NWS of psoriasis patients with normal and decreased saliva secretion was significantly higher than in the control group. Moreover, concentrations of IL-2 (↑20.50%, *p* ≤ 0.0001) and INF-γ (↑18.21%, *p* = 0.005) in NWS of patients with hyposalivation were considerably higher than in psoriasis patients with normal saliva secretion. TNF-α concentration in the NWS of psoriasis patients with hyposalivation was significantly higher than in the control group (↑61.38%, *p* ≤ 0.0001) and in the group of psoriasis patients with normal salivation (↑30.47%, *p* = 0.009). The concentration of IL-10 (↓25.68%, *p* = 0.002; ↓47.30%, *p* ≤ 0.0001, respectively) in the NWS of psoriasis patients with normal and decreased saliva secretion was considerably lower than in the control group. Moreover, the concentration of IL-10 (↓29.09%, *p* = 0.03) in NWS of patients with hyposalivation was significantly lower than in the group of psoriasis patients with normal saliva secretion (Figure 1).

#### 3.1.2. SWS

TNF-α concentration (↑69.77%, *p* ≤ 0.0001; ↑103.32%, *p* ≤ 0.0001, respectively) and IL-2 (↑45.10%, *p* ≤ 0.0001; ↑67.65%, *p* ≤ 0.0001, respectively) in SWS of patients with psoriasis with normal and decreased saliva secretion was significantly higher than in the control group. The concentration of TNF-α (↑19.77%, *p* = 0.0002) and IL-2 (↑15.54%, *p* = 0.03) in SWS of patients with hyposalivation was considerably higher than in psoriasis patients with normal saliva secretion. INF-γ content in SWS of psoriasis patients with hyposalivation was significantly higher than in the control group (↑66.28%, *p* ≤ 0.0001) and in the group of psoriasis patients with normal saliva secretion (↑31.19%, *p* = 0.03). The level of IL-10 (↓44.23%, *p* ≤ 0.0001; ↓61.54%, *p* ≤ 0.0001, respectively) in SWS in psoriasis patients with normal and decreased saliva secretion was significantly lower than in the control group, with IL-10 content (↓31.23%, *p* ≤ 0.0001; ↓61.54%, *p* ≤ 0.0001, respectively) in SWS in hyposalivation patients significantly lower than in psoriasis patients with normal saliva secretion (Figure 1).

**Figure 1.** Cytokine levels in unstimulated and stimulated saliva as well as plasma of plaque psoriasis patients with normal salivation and hyposalivation. Abbreviations: C—the control; IL-2—interleukin 2; IL-10—interleukin-10; INF-γ—interferon-gamma; ns—not significant; NWS—non-stimulated whole saliva; PN—psoriasis patients with normal salivation; PH—psoriasis patients with hyposalivation; SWS—stimulated whole saliva; TNF-α—tumor necrosis factor-alpha. \* *p* < 0.05, \*\* *p* < 0.01, \*\*\* *p* < 0.001, and \*\*\*\* *p* < 0.0001.

#### 3.1.3. Plasma

TNF-α concentration in plasma of psoriasis patients with normal secretion (↑83.25%, *p* ≤ 0.0001) and in plasma of psoriasis patients with hyposalivation (↑100.49%, *p* ≤ 0.0001) was significantly higher than in the control group. Similarly, plasma concentration of IL-2 and INF-γ in psoriasis patients with normal secretion (↑19.68%, *p* = 0.002; ↑28.51%, *p* = 0.0006, respectively) and hyposalivation (↑29.70%, mboxemphp ≤ 0.0001; ↑25.29%, *p* = 0.003, respectively) were significantly higher vs. control. Plasma concentration of IL-10 in psoriasis patients with normal secretion (↓24.49%, *p* ≤ 0.0001) and hyposalivation (↓18.37%, *p* = 0.0001) was significantly lower than in the control group (Figure 1).

#### *3.2. Nitrosative Stress*

#### 3.2.1. NWS

The concentration of NO (↑14.31%, *p* = 0.009; ↑35.14%, *p* ≤ 0.0001, respectively) and nitrotyrosine (↑12.41%, *p* = 0.04; ↑39.60%, *p* ≤ 0.0001, respectively) in NWS of psoriasis patients with normal secretion and with hyposalivation was significantly higher than in the control group. Moreover, the levels of NO (↑18.23%, *p* = 0.0006) and nitrotyrosine (↑24.19%, *p* ≤ 0.0001) in NWS of patients with hyposalivation was considerably higher than in psoriasis patients with normal salivary secretion.

The concentration of S-nitrosothiols and peroxynitrite in NWS of psoriasis patients with hyposalivation was significantly higher than in the control group (↑11.59%, *p* = 0.04; ↑30.70%, *p* ≤ 0.0001, respectively) and the group of psoriasis patients with normal salivation (↑16.64%, *p* = 0.01; ↑17.63%, *p* = 0.003, respectively) (Figure 2).

**Figure 2.** Nitrosative stress in non-stimulated and stimulated saliva as well as plasma of plaque psoriasis patients with normal salivation and hyposalivation. Abbreviations: C—the control; NO—nitric oxide; ns—not significant; NWS—non-stimulated whole saliva; PN—psoriasis patients with normal salivation; PH—psoriasis patients with hyposalivation; SWS—stimulated whole saliva. \* *p* < 0.05, \*\* *p* < 0.01, \*\*\* *p* < 0.001, and \*\*\*\* *p* < 0.0001.

#### 3.2.2. SWS

The concentration of NO, S-nitrosothiols, peroxynitrite, and nitrotyrosine in SWS of psoriasis patients with hyposalivation was significantly higher than in the control group (↑25.56%, *p* = 0.006; ↑35.93%, *p* ≤ 0.0001; ↑11.47%, *p* ≤ 0.0001; ↑11.80%, *p* ≤ 0.0001, respectively), as well as compared to the group of psoriasis patients with normal salivation (↑22.71%, *p* = 0.04; ↑21.53%, *p* = 0.002; ↑17.26%, *p* = 0.002; ↑8.26%, *p* = 0.02, respectively) (Figure 2).

#### 3.2.3. Plasma

The concentration of NO (↑27.19%, *p* = 0.001; ↑33.05%, *p* ≤ 0.0001, respectively) and nitrotyrosine (↑25.04%, *p* ≤ 0.0001; ↑24.34%, *p* ≤ 0.0001, respectively) in plasma of psoriasis patients with normal and decreased saliva secretion was considerably higher than in the control group. Plasma concentration of S-nitrosothiols in psoriasis patients with hyposalivation was significantly higher than in the control group (↑20.41%, *p* = 0.008) and in psoriasis patients with normal saliva secretion (↑24.18%, *p* = 0.008). Plasma concentration of peroxynitrite did not differ between the study and control groups (Figure 2).

#### *3.3. Salivary Gland Function*

Unstimulated as well as stimulated saliva secretion was significantly lower in psoriasis patients with hyposalivation compared to the control group (↓57.58%, *p* ≤ 0.0001; ↓41.03%, *p* ≤ 0.0001, respectively). Similarly, unstimulated as well as stimulated saliva secretion was significantly lower in psoriasis patients with hyposalivation compared to the psoriatic patients with normal salivation (↓56.25%, *p* ≤ 0.0001; ↓34.29%, *p* = 0.0003, respectively). The concentration of protein in NWS of psoriasis patients with hyposalivation was considerably lower than in the control group (↓24.90%, *p* = 0.008). Protein content in SWS of psoriasis patients with hyposalivation was significantly lower than in the controls (↓43.49%, *p* ≤ 0.0001) and the group of patients with normal saliva secretion (↓13.60%, *p* = 0.0008). The activity of salivary amylase in NWS of psoriasis patients with hyposalivation and normal salivation was visibly lower than in the control group (↓30.00%, *p* = 0.0003; ↓25.00%, *p* = 0.002, respectively). Similarly, salivary amylase activity in SWS of psoriasis patients with hyposalivation and normal salivation was significantly lower than in the control group (↓42.86%, *p* ≤ 0.0001; ↓25.00%, *p* = 0.0006, respectively). Moreover, amylase activity in SWS of patients with hyposalivation was considerably lower than in patients with normal salivation (↓23.81%, *p* = 0.02) (Figure 3).

**Figure 3.** Salivary gland function in plaque psoriasis patients and control subjects. Abbreviations: C—the control; NWS—non-stimulated whole saliva; ns—not significant; PN—psoriasis patients with normal salivation; PH—psoriasis patients with hyposalivation; SA—salivary amylase; SWS—stimulated whole saliva; TP—total protein. \* *p* < 0.05, \*\* *p* < 0.01, \*\*\* *p* < 0.001, and \*\*\*\* *p* < 0.0001.

#### *3.4. ROC Analysis*

The assessment of diagnostic usefulness of the analyzed biomarkers of inflammation and nitrosative stress is presented in Tables 2 and 3. Many of the assessed parameters clearly differentiated psoriatic patients with hyposalivation from patients with normal salivary flow. Particularly noteworthy is the assessment of NO, nitrotyrosine, and IL-2 levels in NWS, differentiating psoriatic patients with high sensitivity and specificity on the basis of the rate of saliva secretion (Figure 4).




SWS—stimulated whole saliva; TNF-α—tumor necrosis factor-alpha.

**Figure 4.** Receiver operating characteristic (ROC) analysis of nitric oxide, nitrotyrosine, and IL-2 in unstimulated saliva of plaque psoriasis patients with normal salivation and hyposalivation. IL-2—interleukin 2; NO—nitric oxide; NWS—non-stimulated whole saliva.

#### *3.5. Correlations*

The results of statistically significant correlations are presented in Table 4. We demonstrated a negative correlation between NO concentration and minute secretion of unstimulated saliva as well as between peroxynitrite and protein concentrations in stimulated saliva of patients with hyposalivation. Moreover, we observed a negative correlation between TNF-α level and non-stimulated salivation, as well as in IL-2 content and stimulated salivary flow in patients with hyposalivation. On the other hand, peroxynitrite concentration correlated negatively with α-amylase activity in both unstimulated and stimulated saliva of patients with normal salivation.

We noted a positive correlation between TNF-α and NO concentrations in unstimulated saliva of patients with hyposalivation, as well as between IL-2 and NO contents in unstimulated saliva of psoriasis patients with normal saliva secretion.

We showed a positive correlation between PASI and TNF-α, as well as PASI and IL-2 in unstimulated saliva of patients with hyposalivation. Moreover, we observed a positive correlation between nitrotyrosine concentration and duration of psoriasis in patients with normal as well as reduced salivary flow (both in unstimulated and stimulated saliva).


**Table 4.** Statistically significant correlations in patients with plaque psoriasis.

Abbreviations: IL-2—interleukin 2; NO—nitric oxide; NWS—non-stimulated whole saliva; PN—psoriasis patients with normal salivation; PH—psoriasis patients with hyposalivation; SWS—stimulated whole saliva; TNF-α—tumor necrosis factor-alpha.

#### **4. Discussion**

In the presented study, we evaluated the concentrations of TNF-α, IL-2, INF-γ, IL-10, and selected nitrosative stress parameters (NO, peroxynitrite, S-nitrosothiols, and nitrotyrosine) in NWS and SWS, as well as the plasma of psoriasis patients. The obtained results demonstrated the pathophysiology of salivary gland dysfunction in the course of plaque psoriasis. We were also seeking salivary psoriasis biomarkers that could be helpful in diagnosing the severity of psoriasis and its salivary complications.

The accepted values of normal unstimulated salivary flow are above 0.2 mL/min. Any unstimulated flow rate below 0.2 mL/min is considered salivary gland hypofunction and is referred to as hyposalivation [18]. Hyposalivation has a detrimental effect on numerous aspects of oral health, and thus on general well-being. It decreases the quality of life as it hinders speaking, tasting, chewing, and swallowing food [40]. Reduced saliva secretion is the cause of cracks and fissures in the oral mucosa, which is associated with chronic pain in the oral cavity and the resulting discomfort to the patient. Decreased salivation also contributes to boosted incidence of caries, periodontitis, and fungal infections of the oral cavity [40]. All these may lead to patient malnutrition, social isolation, and even depression, as well as generate high treatment costs. Therefore, it is very important to identify patients with salivation disorders and to prevent the development and effects of salivary gland dysfunction in the course of systemic diseases.

Our findings that plasma concentrations of TNF-α, IL-2, and INF- γ were significantly higher and for IL-10 were significantly lower than in the controls are consistent with the assumption that psoriasis is primarily driven by an aberrant immune response and results from an imbalance between Th1 and Th2 cells [2–4]. Significantly higher concentration of NO, especially nitrotyrosine, and a positive correlation between the latter and the disease duration in psoriatic patients' plasma compared to the controls confirm the contribution of nitrosative stress to the development of the disease [41]. Interestingly, apart from S-nitrosothiols, we did not observe significant differences between patients with hyposalivation and those with normal saliva secretion. On the other hand, patients from the hyposalivation group were characterized by longer duration of psoriasis and higher PASI index compared to those with normal flow of unstimulated saliva.

At this point, it is worth reminding that 90% of saliva is produced by three pairs of large salivary glands: submandibular, parotid, and sublingual. The remaining 10% of saliva is secreted by small salivary glands scattered under the oral cavity mucosa, and gingiva fluid. The submandibular glands are the major contributor to unstimulated salivary flow, and the parotid glands secrete stimulated saliva, that is, saliva secreted mainly in response to stimuli. The contribution of the sublingual glands to unstimulated and stimulated salivation is low [42]. Therefore, any deviation in the composition of unstimulated saliva reflects dysfunction of the submandibular glands, as well as of stimulated saliva, of the parotid glands. The exception here are patients with inflammatory changes in periodontal tissues, in whom changes in saliva composition reflect periodontal diseases. In our work, we excluded patients with periodontitis/gingivitis, and therefore any changes observed in saliva originated from the dysfunction of salivary glands.

Our results revealed significantly increased levels of the tested proinflammatory cytokines and a decrease in IL-10 concentration in unstimulated and stimulated saliva of psoriasis patients with normal salivation (except for TNF-α in NWS and INF-γ in SWS) and hyposalivation compared to the controls. An earlier report suggested that mRNA expression of Th1 and derived inflammatory cytokines IL-2, TNF-α, and INF-γ were also increased in the saliva of the patients with Sjögren's syndrome [43]. The salivary changes were accompanied by clusters of infiltrating cells present in salivary gland biopsy, where 80% were Th1 cells, and the remaining 20% consisted of stimulated B lymphocytes and plasma cells [43]. On the basis of the performed analyses, it was difficult to assess the nature of the developing inflammation in salivary glands of our patients. There were also no histological examinations of the salivary glands of psoriasis patients. By analogy to Sjögren's syndrome, the observed increases in TNF-α, IL-2, and INF-γ concentrations allow us to assume that salivary glands of patients with psoriasis are infested with autoreactive Th1 lymphocytes. Despite the deficiency of Th2 response (↓IL-10) supporting the humoral type response, we do not rule out the presence of stimulated B lymphocytes, as there has been no research to confirm or exclude the existence of the culprit autoantigens. However, increased concentration of the examined proinflammatory cytokines and decreased level of IL-10 in NWS and SWS of patients with hyposalivation compared to psoriasis patients with normal salivation indicates an increase in imbalance between Th1 and Th2 cells, and thus inflammation in salivary glands of patients with hyposalivation vs. those with normal salivary flow.

Human salivary glands contain different kinds of nitric oxide synthase (NOS) isoforms. Neuronal NOS (*n*NOS) was found in the salivary gland parenchyma, ducts, blood vessels, and nerve fibers around acini, mainly in the submandibular glands [44], and—in negligible amounts—in the parotid and sublingual glands [45]. Endothelial NOS (*e*NOS) was identified as localizing to the glandular vascular endothelium of the salivary ducts [45]. *i*NOS has been detected in the salivary ducts of normal tissue [45]. In physiological concentrations, NO does not damage the structures of salivary glands; it regulates oral blood flow and saliva secretion, and participates in non-specific protective mechanisms [23–25], which seems to take place in parotid glands of psoriasis patients without salivation disorders (no changes in the studied nitrosative stress parameters in SWS). We observed excessive amount of NO and peroxynitrite in unstimulated and stimulated saliva of patients with hyposalivation. A positive correlation between TNF-α and NO concentrations in NWS of patients with hyposalivation and between IL-2 and NO content in NWS of patients with normal salivation confirm the previous observations that proinflammatory cytokines lead to the expression of *i*NOS in salivary gland cells, resulting in increased production of NO and its derivatives [46]. We also noted a boost in nitrosative stress (NO, S-nitrosothiols, peroxynitrite) and, primarily, nitrosative damage to protein elements of the salivary glands (S-nitrosothiols and nitrotyrosine) in unstimulated and stimulated saliva of psoriasis patients with hyposalivation vs. those with normal saliva secretion.

Evidence has shown that much larger amounts of NO generated in response to inflammation are connected with the cytotoxic effect of NO due to its interaction with superoxide anions to form peroxynitrite and other free radicals. Research results have revealed that intense production of NO and peroxynitrite in salivary glands acts as a strong proapoptotic agent [46,47]. Moreover, it has been observed that NO, by auto-ADP (adenosyno-diphospate) ribosylation of glyceraldehyde 3-phosphate dehydrogenase, inhibits the production of ATP that is necessary to maintain anabolic processes in the cell [23]. It has been demonstrated that apoptosis of salivary gland structures disturbs their function, and ATP deficiency impairs mechanisms responsible for replacing damaged or lost cellular elements [46,48]. We noted a negative correlation between NO and NWS secretion, and between peroxynitrite and protein concentrations in SWS of hyposalivation patients. These results suggest that decreased salivary secretion and protein synthesis/selection could be caused by the proapoptotic effect of NO on the salivary gland cells. This hypothesis requires further confirmation in histological studies. On the other hand, it is known that TNF-α and IL-2 stimulate the production of metalloproteinases, which results in structural changes in the basement membrane of the salivary glands [49]. A negative correlation between TNF-α and NWS secretion, and IL-2 and SWS secretion in patients with hyposalivation may result from damage to acinar cell-basement membrane interaction resulting from overproduction of MMPs (metalloproteinases) followed by decreased number of secretory units (acini and ducts) [50,51]. This phenomenon has been recently demonstrated in the saliva of Sjögren's syndrome patients [52]. Remodeling of the extracellular matrix, alongside apoptosis, could be the reason for the observed drop in the synthesis/secretion of proteins and reduced salivary secretion in psoriasis patients with hyposalivation. It is noteworthy that salivary gland dysfunction occurs in patients with a longer duration and higher intensity of the disease.

The lack of significant differences in the secretion of NWS and SWS as well as proteins between patients with normal salivation (shorter disease duration, lower PASI) and the controls suggests that, at an early stage of the disease, the mechanisms of controlling saliva secretion and protein production/secretion counteract the damaging effects of psoriasis. Interestingly, already at this early stage we observed decreased amylase activity in SWS and UWS of patients with normal salivation, as well as intensification of this phenomenon in the saliva of patients with hyposalivation. These results may explain the negative correlation between peroxynitrite concentration and salivary amylase activity in NWS and SWS of patients with normal salivation. It has been demonstrated that peroxynitrite reacts readily with iron-sulfur cluster of several enzymes and is able to oxidize the sulfhydryl groups of proteins, leading to the formation of disulfides and resulting in their inactivation [53]. Naturally, it should be remembered that exposure to peroxynitrite entails tyrosine nitration of proteins [54]. This mechanism of amylase inactivation should be eliminated, as increased nitrotyrosine concentration was only observed in NWS and SWS of patients with hyposalivation, both compared to the controls and patients with normal saliva secretion.

Salivary glands are surrounded by a dense network of blood vessels that enable the exchange of components between the acinar cells and ducts, as well as blood. Thus, biomarkers present in the blood can permeate into the structures of salivary glands and hence into saliva. Therefore, saliva is more and more frequently considered a potential source of biological markers for systemic diseases. Many of the examined parameters clearly differentiated psoriatic patients with hyposalivation from psoriatic patients with normal saliva flow, and thus the levels of NO, nitrotyrosine, and IL-2 in NWS deserve special attention and should be further evaluated. Additionally, the observed positive correlation of PASI and TNF-α and IL-1β in UWS of patients with hyposalivation could provide a new non-invasive and simple method in the diagnosis of the intensity of the disease.

#### **5. Conclusions**

Increased levels of TNF-α, IL-2, and INF-γ, as well as decreased content of IL-10 in NWS and SWS of psoriasis patients compared to the controls indicated an imbalance between Th1 and Th2 cells in the salivary glands.

The severity of inflammation and nitrosative stress in the salivary glands of psoriatic patients depends on the disease duration.

At an early stage of the disease, the mechanisms controlling saliva secretion and protein production/secretion counteract the damaging effects of psoriasis. With the severity and duration of psoriasis, the secretory function of all salivary glands is lost, which is manifested as significant reduction of unstimulated and stimulated saliva secretion as well as protein concentration.

Dysfunction of salivary glands in the course of psoriasis may be attributed to inflammation and nitrosative stress.

**Author Contributions:** Conceptualization, A.Z., M.M.; data curation, A.S.-R., U.K., A.K., A.K.-R., J.K.; formal analysis, A.Z., M.M., A.S.-R.; funding acquisition, A.S.-R., A.Z.; investigation, A.S.-R., A.Z., M.M.; methodology, A.Z., M.M.; material collection: A.S.-R., U.K., A.K., A.K.-R.; supervision, A.Z., I.F., J.K.; validation, A.S.-R., A.Z., M.M.; visualization, M.M.; writing—original draft, A.S.-R., A.Z., M.M.; writing—review and editing, A.Z., M.M. All authors have read and agreed to the published version of the manuscript.

**Funding:** This work was supported by grants from the Medical University of Bialystok, Poland (grant numbers: SUB/1/DN/20/002/1209; SUB/1/DN/20/002/3330).

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

*Article*
